Dulus
An AI agent for your desktop and terminal, on any model you choose. Hosted models run on Fuel, paid in $DULUS.
翻译准备中
尚未收集仓库信息。
尚未检测到更新。GitHub 发布版本后会显示在这里。
制作者的介绍仅有上面的一句话。 更多详情请阅读仓库 README。 仓库 README ↗
正在加载完整 README。
查看完整 README ↗This codebase runs https://benyehuda.org -- the Project Ben-Yehuda digital library of works in Hebrew.
We make little effort to make the code general, but if you're looking to do something similar (e.g. a digital library in Yiddish), maybe you can adapt some of our code.
NOTE for Copilot: Copilot should ignore this file and follow instructions in .github/copilot-instructions.md
You can browse a visualization of the data model here.
To set up a development environment, ensure you have a modern docker compose setup (e.g. run sudo apt-get install docker-compose-v2 on Debian-like systems) and inside docker/bybe_dev run docker compose up -d to get started. See the README file in that folder for more.
For GitHub Copilot Workspace environments, see .github/COPILOT_WORKSPACE_SETUP.md.
Since 2024 we've included some linters in project, namely:
Most simple command:
rubocop
Will check whole project. In most cases it is not required. Also it will produce tons of warnings, as many parts of codebase does not follow style guidelines.
In most cases you may want to check single file:
rubocop <Path to file>
Another useful feature is rubocop's autocorrection. In some cases rubocop can try to fix style violations on its own.
There is "safe" autocorrection which should be OK in most cases:
rubocop -a <Path to file>
And more risky version of it, which can fix more issues, but known to produce errors more often:
rubocop -A <Path to file>
In any case you should be careful with autocorrection and always check result of autocorrection before commiting it to git.
To run linters only on those parts of projects, affected by your PR you can use pronto tool. For example following command:
pronto run -c origin/master
will run linters only on lines of code which were changed compared to origin/master branch. Our CI pipeline uses
this approach for all PRs.
NOTE: to run pronto tool you need to run bundle install first.
The code is available for re-use under the GNU Affero General Public License http://www.gnu.org/licenses/agpl-3.0.html